Based on the results of your test, could it be possible that the discolored chest is a *fake* one? Perhaps locating that one first can increase your chance from 33% to 50%. Probably needs more testing first though. --Toksyuryel 13:43, 25 January 2007 (EST)
I've done this a few times and have never been able to see any differences in the colors. It also never really make sense to me that you would be able to see a difference. There is no reason why the "roll" wouldn't happen when you select a chest. At that moment the computer rolls, 1-666 and a mimic spawns, 667-999 and a chest spawns. --Gahoo 14:03, 25 January 2007 (EST)
Well, the whole basis for the percentage chances were that it's assumed the correct chest is determined when you enter. If in fact it is randomly determined only once you pick a chest, then the percentage chance is likely to be very different. This opens a whole new can of worms now... --Toksyuryel 15:19, 25 January 2007 (EST)
Indeed.. it is -not- a 33% chance. Or, to be more accurate, I am 99% certain it is not a 1/3 chance, and 96% certain it is not a 1/4 chance. Being math oriented, decided to take a systematic approach. Before now, I had done this bcnm between 2 or 5 times . Today, I did this BCNM 10 times in a row, selecting the small box each time (so as to minimize differences between trials). Attempts 1-5 were mimics, attempt 6 was an armoury chest, and attempts 7-10 were all mimics. If I was to make a guess on this alone, it would be reasonable to assume that there is a 1/8 chance of being a chest... and it would be unreasonable on the other end to assume it is a less than 1/20 probability of a chest. Anyone that wants to continue this is welcome to, but 600+ beastman seals is enough for me ^_^
Runeghost 00:07, 6 May 2007 (CDT)
Unfortunately, 10 times in a row is not nearly enough of a test sample to qualify as a valid test. I've had BCNM runs that are nothing but bad luck 15 times in a row, and I've also had runs where we've gotten the "awesome item" every time. I've only ever done this BCNM once, so I have no personal involvement in testing, but I chose the small chest, and I won (Just lucky that time I guess). After that, I didn't want to press my luck, so I haven't spent (gambled) anymore of my seals on this one. You have to really have a large community of people, reporting their loot/success rate every time to start getting a good grip on the odds of getting an item, or in this case, the chance to just win without a fight. --Chrisjander 00:39, 6 May 2007 (CDT)
Oh, I understand. I am not trying to say "it is a 1/8" chance... and good luck/bad luck, streaks, etc are a part of the game. Alright, lets assume that there is a fixed percentage of 'winning', assume p=1/3. In that case, over 10 trials, the expected outcome would be =np=10*(1/3)=3.33 etc. The variance would then be (n*p*(1-p))^.5, (10*1/3*2/3)^.5,or 1.49 (roughly). Now, I was assuming when I did the numbers above that I had done it 2-5 times previous, lets say 2, so expected outcome=12*1/3=4, and variance=1.6.. so I suppose my one win over 12 trials is only about 2 sigmas out. So I -could- have had a bad day, but if I remember my confidence intervals, more than 96% of all tests should be withing 2 sigmas of the expected mean. Again, I ran out of beastman seals, but if anyone else would like to take 5 or 10 or 20 attempts in a row and provide data, I will happily add it to my numbers. To summarize, I am not trying to say what the probability -is-, just that I don't think it is 1/3. Runeghost 16:38, 6 May 2007 (CDT)
Alright, finally got a few more stacks of Beastman seals and decided to take another crack at this, this time, eighteen attempts, 3 wins <For the record, 3x mithril ore, 3x Soldier's Rings -_-;;; > Maybe in another year or so I will make another very ..very long day of pouring seals down the drain =)
Runeghost 05:17, 15 May 2008 (UTC)
"Do NOT try to use any type of warp ring such as Tavnazian Ring or Olduum Ring as they do not work in BCNM areas. Tested in Treasure and Tribulations, ring would not activate after Perfect Dodge when finding the Mimic, and died." Hm... Level 50 BCNM.. level 60 rings.... Yea that makes sence. You wouldn't have been able to equip these at the time anyway and even now with level sync the enchantment still won't work. Natsuchii 06:54, 2 November 2008 (UTC)
So after reading the strats for this BCNM fight, I thought of something no one else seemed to have mentioned; "What if my Automaton attacked?" Automatons on RNG and BLM frame that fire a ranged attack or start casting with a nuke spell do not get claim on a mob; does not change the mob's name to red. Avatars and Call Beast pets attack and gain claim on a mob; or you tell the avatar to use a Blood Pact and you aggro the mob. PUP you can Deploy > Deactivate and lose all hate. So, with this in mind, I decided to go test this out. I decided to use Sharpshot frame for the tests; six runs were all I had seals for at the time. The first three runs, I always chose the Large box. All three were "correct". For the last three, I switched between each box; Small > Medium > Large. Each time a shot was fired off, all three boxes vanished and the Armoury Crate was left. Sure, you could say that going 100% on guessing the correct chest is just luck, but how often has that happend in previous testing? And, for that matter, the first three were all Large Box. Exactly what are the odds of that happening in previous tests? If anyone else would care to test this, feel free to. I'm certain you'll get the same results with this method. Mizuharu 03:28, 22 March 2010 (CST)